Multiomic Immunophenotyping of COVID-19 Patients Reveals Early Infection Trajectories

Yapeng Su, Daniel Chen, Christopher Lausted, Dan Yuan, Jongchan Choi, Cheng Dai, Valentin Voillet, Kelsey Scherler, Pamela Troisch, Venkata R. Duvvuri, Priyanka Baloni, Guangrong Qin, Brett Smith, Sergey A. Kornilov, Clifford Rostomily, Fan Xu, Jing Li, Shen Dong, Alissa C. Rothchild, Jing Zhou, Kim M. Murray, Rick Edmark, Sunga Hong, Lesley Jones, Yong Zhou, Ryan Roper, Sean Mackay, D. Shane O’Mahony, Christopher Dale, Julie A. Wallick, Heather A. Algren, Michael Zager, Andrew T. Magis, Wei Wei, Nathan D. Price, Sui Huang, Naeha Subramanian, Kai Wang, Jennifer Hadlock, Leroy Hood, Alan Aderem, Jeffrey A. Bluestone, Lewis L. Lanier, Phil Greenberg, Raphaël Gottardo, Mark M. Davis, Jason D. Goldman, James R. Heath, the ISB-Swedish Covid19 Biobanking UnitbioRxiv 2020 · DOI

Abstract

Host immune responses are central to the control and progression of SARS-CoV-2 infection, yet the complexity of these responses remains incompletely understood. This study presents a comprehensive immune response map integrating 454 plasma proteins and 847 metabolites with single-cell multi-omic assays of peripheral blood mononuclear cells (PBMCs). By co-analyzing whole transcriptomes, 192 surface proteins, and T and B cell receptor sequences alongside clinical measures from 50 patient samples, we identify novel cellular subpopulations—including proliferative exhausted CD8+ and CD4+ T cells and cytotoxic CD4+ T cells—associated with severe disease. We condensed over one million immune features into a single "M1" immune response axis that aligns with clinical severity and inflammatory markers, providing a high-resolution resource for understanding COVID-19 pathogenesis and informing therapeutic strategies.

Introduction

The clinical presentation of COVID-19 is notoriously heterogeneous. While many individuals experience mild symptoms, others suffer from fatal cytokine storms, multi-organ failure, and prolonged hospitalization. To understand this variance, the immune system must be viewed not as a collection of isolated cells, but as a finely tuned biological orchestra. Previous research has often focused on single modalities, such as antibody titers or gene expression in isolation. This study addresses the need for a unified, longitudinal view that connects clinical symptoms with the underlying molecular "chatter" between the innate and adaptive immune systems during the critical first week of hospitalization.

A Multi-omic Systems Approach

To capture the full breadth of the host response, researchers employed a multi-omic framework, measuring multiple types of biological molecules simultaneously from the same patient samples. The study utilized a longitudinal cohort from the ISB-Swedish COVID19 Biobanking Unit, consisting of 26 hospitalized patients (providing 50 blood samples across two time points) and 35 healthy controls.

Methodology Overview

  • Single-Cell Multi-omics: Using 10X Genomics and CITE-seq, the team analyzed the whole transcriptome and 192 surface proteins in individual PBMCs.
  • Immune Repertoire: T-cell and B-cell receptor (TCR/BCR) sequencing was used to track the expansion of specific immune lineages.
  • Single-Cell Secretome: The Isoplexis platform measured 32 secreted cytokines from individual cells to map functional communication.
  • Plasma Analytics: 454 proteins (Olink) and 847 metabolites (Metabolon) were quantified to provide a systemic view of the blood environment.
  • Surprisal Analysis: This mathematical technique was used to condense millions of variables into a single "M1" immune response axis, representing the coordinated shift of the immune system away from a healthy baseline.

The M1 Axis: A Molecular Mirror of Severity

The most striking finding of the study is that the complexity of the immune response can be captured by a single mathematical axis. The M1 score showed a strong Spearman correlation (0.58, p < 0.001) with the WHO ordinal scale of disease severity. High M1 scores aligned independently with clinical inflammatory markers such as C-reactive protein (CRP), D-dimer, and Ferritin, as well as the clinical requirement for mechanical ventilation. This suggests that the molecular state of a patient's blood is a direct and quantifiable mirror of their physical condition.

Cellular Hallmarks of Severe Infection

The Exhaustion Paradox

The researchers identified a novel subpopulation of CD8+ T cells (Cluster 8) characterized by a paradoxical state: they are simultaneously "exhausted" (expressing markers like LAG3 and TIGIT) and "proliferative" (expressing MKI67). While exhausted cells typically stop dividing, in severe COVID-19, these cells appear trapped in a frantic but dysfunctional cycle of activation.

Killer "Helper" Cells

CD4+ T cells are traditionally "helpers" that coordinate the immune response. However, in severe cases, the study identified cytotoxic CD4+ T cells. These cells develop "killer" functions usually reserved for CD8+ cells, likely a result of repeated, intense viral encounters that force a functional shift in the T-cell compartment.

Monocyte Immunoparalysis

In critically ill patients (ICU-admitted), monocytes exhibited a robust downregulation of MHC-class II molecules. This state, termed "immunoparalysis," is similar to what is observed in sepsis. It suggests that the innate immune system loses its ability to present viral antigens to the adaptive system, effectively blinding the body's defenses at a critical juncture.

Metabolic Shifting and Clinical Trajectories

The study noted a sharp increase in kynurenine and a corresponding drop in tryptophan in severe cases. This shift suggests the activation of the IDO enzyme, which is known to contribute to hypotension (low blood pressure)—a dangerous symptom in the most critically ill patients.

By tracking patients between the first (T1) and second (T2) time points, the researchers identified molecular signatures of recovery. Patients who improved showed a measurable decrease in Type I interferon signaling and T-cell activation markers, providing a template for predicting patient trajectories.

Conclusion and Future Directions

While the cohort of 26 patients is modest, the depth of data per patient provides a high-resolution map of the early infection trajectory. The study identifies specific pathways—such as the IDO enzyme or the expansion of plasmablasts and proinflammatory monocytes—that could serve as targets for future therapies. This systems immunology framework establishes a new standard for studying infectious diseases, moving beyond single-metric observations toward a holistic understanding of host-pathogen dynamics.

Summary updated

Abstract

Host immune responses play central roles in controlling SARS-CoV2 infection, yet remain incompletely characterized and understood. Here, we present a comprehensive immune response map spanning 454 proteins and 847 metabolites in plasma integrated with single-cell multi-omic assays of PBMCs in which whole transcriptome, 192 surface proteins, and T and B cell receptor sequence were co-analyzed within the context of clinical measures from 50 COVID19 patient samples. Our study reveals novel cellular subpopulations, such as proliferative exhausted CD8 <sup>+</sup> and CD4 <sup>+</sup> T cells, and cytotoxic CD4 <sup>+</sup> T cells, that may be features of severe COVID-19 infection. We condensed over 1 million immune features into a single immune response axis that independently aligns with many clinical features and is also strongly associated with disease severity. Our study represents an important resource towards understanding the heterogeneous immune responses of COVID-19 patients and may provide key information for informing therapeutic development.
BibTeX
@misc{su2020multiomic,
  author = {Yapeng Su and Daniel Chen and Christopher Lausted and Dan Yuan and Jongchan Choi and Cheng Dai and Valentin Voillet and Kelsey Scherler and Pamela Troisch and Venkata R. Duvvuri and Priyanka Baloni and Guangrong Qin and Brett Smith and Sergey A. Kornilov and Clifford Rostomily and Fan Xu and Jing Li and Shen Dong and Alissa C. Rothchild and Jing Zhou and Kim M. Murray and Rick Edmark and Sunga Hong and Lesley Jones and Yong Zhou and Ryan Roper and Sean Mackay and D. Shane O’Mahony and Christopher Dale and Julie A. Wallick and Heather A. Algren and Michael Zager and Andrew T. Magis and Wei Wei and Nathan D. Price and Sui Huang and Naeha Subramanian and Kai Wang and Jennifer Hadlock and Leroy Hood and Alan Aderem and Jeffrey A. Bluestone and Lewis L. Lanier and Phil Greenberg and Raphaël Gottardo and Mark M. Davis and Jason D. Goldman and James R. Heath and the ISB-Swedish COVID19 Biobanking Unit},
  title = {{Multiomic Immunophenotyping of COVID-19 Patients Reveals Early Infection Trajectories}},
  year = {2020},
  doi = {10.1101/2020.07.27.224063},
  howpublished = {bioRxiv},
}
Figure